The distinction between the concepts data, information and knowledge is a key component of the conceptual scaffolding of the discipline of knowledge management. In this paper we have sought to demonstrate and solve a series of major problems relating to the distinction between these concepts. We have first reviewed one version of the distinction (the "standard" model), showing the degree of confusion that surrounds it, and propose an alternative distinction. We then argue that there is a critical operation on which the successful execution of a knowledge management program to a large extent depends, namely the identification of knowledge. We propose that, in order to be able to make that identification, we need a definition of knowledge that is conceptually correct and pragmatically useful. While there are a number of theoretical models that could meet this objective, there is one in particular that has traditionally been held up as being solid and of broad explanatory range, i.e. that of classical epistemology. However, deeper analysis of this proposal brings us to very different conclusions. For one thing, the definition of knowledge it puts forward has a number of powerful limitations of a conceptual nature (being unable to deal with the lottery paradox or the counter-examples of the epistemologist Edmund Gettier) that tend to invalidate it as a proposal. And, on the other hand, there are also major pragmatic limitations to its definition of knowledge: it is limited when it comes to identifying tacit knowledge and very poor ecologically, requiring major investment (principally metacognitive in nature) of its potential users.

Knowledge management, data, information, identification of knowledge, classical epistemology

Knowledge Assets Identification and Methodologies of Implementation in Organizational Knowledge Management (KAIMI)